fiscal decentralization and economic growth: a cross-country study

Ž .JOURNAL OF URBAN ECONOMICS 43, 244]257 1998ARTICLE NO. UE972042

Fiscal Decentralization and Economic Growth:A Cross-Country Study*

Hamid Davoodi and Heng-fu Zou

Policy Research Department, The World Bank, N10-075, 1818 H St. NW, Washington, DC20433; and Institute of Ad¨anced Studies, Wuhan Uni ersity, Wuhan, China 430072

Received May 14, 1996; revised January 31, 1997

We use a panel data set of 46 countries over the 1970]1989 period to investigatethe relationship between fiscal decentralization and economic growth. We find anegative relationship between fiscal decentralization and growth in developingcountries, but none in developed countries. Several explanations are offered forour findings. Q 1998 Academic Press

1. INTRODUCTION

Out of the seventy-five developing and transitional economies withpopulations greater than five million, all but twelve claim to have em-

Žbarked on some type of transfer of power to local governments Dillingerw x.7 . Fiscal decentralization, or the devolution of fiscal power from thenational government to subnational governments, is seen as part of areform package to improve efficiency in the public sector, to increasecompetition among subnational governments in delivering public services,

Ž w xand to stimulate economic growth Bahl and Linn 1 , Bird and Wallichw x.4 .

The basic economic argument in favor of fiscal decentralization is basedŽ .on two complementary assumptions: 1 decentralization will increase

economic efficiency because local governments are better positioned thanthe national government to deliver public services as a result of informa-

Ž .tion advantage; and 2 population mobility and competition among localgovernments for delivery of public services will ensure the matching of

Ž w xpreferences of local communities and local governments Tiebout 15 ;

*For criticisms, suggestions and help, we thank Richard Bird, Jan Brueckner, ShantayananDevarajan, Andrew Feltenstein, Avner Greif, Bert Hofman, Gregory Ingram, Ronald McKin-non, Charles McLure, Wallace Oates, Yingyi Qian, Anwar Shah, Hedy Sladovich, BarryWeingast, Danyang Xie, Tao Zhang, and seminar participants at Stanford University, WuhanUniversity, and the World Bank. We are most grateful to two referees and Jan Brueckner fortheir detailed suggestions, which led to a substantial revision of this paper.

244

0094-1190r98 $25.00Copyright Q 1998 by Academic PressAll rights of reproduction in any form reserved.

FISCAL DECENTRALIZATION AND GROWTH 245

w x.Oates 11 . These public-finance considerations suggest that policies aimedat the provision of public services such as infrastructure and education thatare sensitive to regional and local conditions are likely to be more effectivein encouraging growth than centrally-determined policies that ignore thesegeographical differences. Consequently, other things being equal, a decen-tralized fiscal system where local governments play a more important rolethan the federal or central government in public-service provision leads to

Ž w x.more rapid economic growth Oates 13 .Although many policy discussions have favored decentralization, there is

little empirical support to the hypotheses mentioned above. The objectivesof our study are to supply an analytical framework and empirical method-ology, and to use the methodology to test for the presence and size ofefficiency gains from fiscal decentralization.

Fiscal decentralization is a complicated phenomenon with many dimen-sions. This paper will focus on one important dimension: economic growth.Section 2 provides a tractable theoretical and empirical framework linkingfiscal decentralization to growth, and characterizes parameters that mea-sure the efficiency gains from fiscal decentralization. The growth dimen-sion of fiscal decentralization is emphasized for two reasons. First, eco-nomic growth is often cited as a major objective of fiscal decentralizationŽ w x w x w x.Bahl and Linn 1 , Bird and Wallich 4 , Oates 13 . Second, an often-statedobjective of many governments is to adopt policies that lead to a sustainedincrease in per capita income. In that context, it is important to know

Ž .which level of government national or subnational contributes more toeconomic growth.

Section 3 provides a detailed empirical examination of the relationshipbetween fiscal decentralization and economic growth. Section 4 concludesand points to some limitations of the study.

2. ANALYTICAL FRAMEWORK

In this section, we outline a theoretical model of fiscal decentralizationand economic growth. The model assumes, without loss of generality, threelevels of government: federal, state, and local. The level of fiscal decentral-ization is defined as the spending by subnational governments as a fractionof total government spending. For example, fiscal decentralization in-creases if spending by state and local governments rises relative to spend-ing by the federal government.

w xFollowing Barro 3 , the production function has two inputs: privatecapital and public spending. We depart from the Barro model by assumingthat public spending is carried out by three levels of government: federal,state, and local. Let k be private capital stock, g total governmentspending, f federal government spending, s state government spending,

DAVOODI AND ZOU246

and l local government spending, all measured on a per capita basis:

f q s q l s g . 2.1Ž .

The production function is Cobb]Douglas1:

y s k a f bsg l v , 2.2Ž .

where y is per capita output, 1 ) a ) 0, 1 ) b ) 0, 1 ) g ) 0, 1 ) v ) 0,and a q b q g q v s 1.

The allocation of consolidated or total government spending g amongdifferent levels of government takes the following form:

f s u g , s s u g , l s u g , 2.3Ž .f s l

where u q u q u s 1 and 0 - u - 1 for i s f , s, and l. Thus, u is thef s l i fshare of federal government in total spending, u the share of statesgovernment, and u the share of local government. Consolidated govern-lment spending g is financed by a flat income tax at rate t :

g s t y. 2.4Ž .

The representative agent’s preferences are given by

`1ysc y 1

yr tU s e dt , 2.5Ž .H 1 y s0

where c is per capita private consumption, and r is the positive timediscount rate.

The dynamic budget constraint of the representative agent is

dka b g vs 1 y t y y c s 1 y t k f s l y c. 2.6Ž . Ž . Ž .

dt

We further assume a constant tax rate along the balanced growth path.Given total government spending g, a constant tax rate t , and the shares

Ž .of spending by different levels of governments u ’s, i s f , s, l , the repre-iŽ .sentative agent’s choice of consumption is determined by maximizing 2.5

Ž .subject to 2.6 and the government’s budget allocation. Along the bal-anced growth path, the solution for the per capita growth rate of the

1The use of more general functional forms such as the CES would not alter our analysisw x w xqualitatively; see Davoodi, Xie, and Zou 5 and Devarajan, Swaroop, and Zou 6 .


economy is given by

dyrdt 11ya ra b ra g ra v ras 1 y t t au u u y r . 2.7Ž . Ž .f s ly s

Ž .2Equation 2.7 shows that the long-run growth rate of per capita outputis a function of the tax rate and the shares of spending by different levelsof government. It forms the basis for our empirical investigation of therelationship between fiscal decentralization and growth. Following theliterature on fiscal federalism, we regard a country as more fiscallycentralized if it has a higher value of the federal spending share u .f

It is important to note that, for a given share of total governmentspending in GDP, a reallocation of public spending among different levelsof governments can lead to higher economic growth if the existing alloca-tion is different from the growth-maximizing expenditure shares. To show

Ž .this point, we maximize the growth rate in 2.7 by choosing u , u , and uf s lsubject to the constraint u q u q u s 1. The growth-maximizing govern-f s lment budget shares are

b g vU U Uu s , u s , u s .f s lb q g q v b q g q v b q g q v

Therefore, as long as the actual government budget shares are differentfrom growth-maximizing shares, the growth rate can always be increasedwithout altering the total budget’s share in GDP.

3. EMPIRICAL ANALYSIS

3.1. Econometric Specification and Data SourcesŽ .Equation 3.1 is the growth regression that will be estimated on a

cross-country panel data using the ordinary least squares technique:

g s d q d u q d t q d X D q d X N q d X X q « , 3.1Ž .i t 1 2 i t 3 i t 4 i 5 t 6 i t i t

Ž . Ž .where i s 1, . . . , I and t s 1, . . . , N refer to country i at time t; Idenotes the number of countries and N the number of time periods; d ,1d , and d are scalar parameters while d X , d X and d X are vectors; g is the2 3 4 5 6 i taverage growth rate; u is the measure of fiscal decentralization; t is thei t i t

Žtax rate; D is a vector of I y 1 country fixed-effects i.e., country dum-i. Žmies ; N is a vector of N y 1 time fixed-effects i.e., intercept timet

.dummies . We work with time-averaged data since the benefits of fiscal

2 w xSee Davoodi, Xie, and Zou 5 for a more general expression of the balanced growth ratewith the CES production technology.

DAVOODI AND ZOU248

decentralization are not expected to affect year-to-year fluctuations ingrowth. Because of our focus on long-run growth, the growth regression isestimated on data averaged over five- and ten-year periods.3 Accordingly,the dependent variable is the average growth rate over these two periods;X is a vector of control variables; and « is the disturbance term that isi t i t

assumed to be serially uncorrelated and orthogonal to the explanatoryŽ .variables. Our primary concern is the coefficient d on the fiscal decen-2

tralization variable, which is expected to be positive and significant giventhe conventional arguments in favor of fiscal decentralization.

The average growth rate is the average growth of real per capita outputover five- and ten-year periods. Real per capita output is the real per

Ž .capita gross domestic product GDP at 1985 international prices and isw x Ž .taken from the Summers]Heston 14 data set version 5.6a . The tax rate

is the ratio of total tax revenues to GDP, both in nominal terms and inlocal currency; these variables are taken from the International Monetary

Ž .Fund’s Government Finance Statistics GFS and the World Bank Eco-Ž .nomic and Social Data BESD base, respectively.

The measure of fiscal decentralization is the subnational share of totalgovernment spending. The higher is this measure, the higher is the degreeof fiscal decentralization. Such a measure has been constructed previously

w xby, inter alia, Oates 12, 13 . The numerator of the fiscal decentralizationvariable is direct spending by subnational governments, i.e., their totalspending net of intergovernmental transfers. The denominator is the sum

Žof spending by the national government i.e., the consolidated central. Ž .government and subnational governments state and local net of inter-

governmental transfers. The GFS is the primary source for internationallycomparable data on economic activities at all levels of government. Toincrease the sample size, for countries with three levels of government we

Žhave consolidated accounts of the two subnational governments state and.local governments into one, thus enabling us to pool the fiscal decentral-

ization measure for these countries with countries that have only onesubnational government.

The vector X consists of a set of variables identified by Levine andi tw xRenelt 9 as the important control variables for cross-country growth

Ž . Ž .regressions. These are i the average growth rate of population; ii initialŽ . Ž .human capital; iii initial per capita GDP; and iv the average real

investment share of GDP. The first two variables are taken from theWorld Bank’s BESD; the latter two are from the Summers]Heston data

3 w xSee Barro and Sala-i-Martin 2 for cross-country growth regressions on five- and ten-yearaverage data.


base. The measure of human capital is the secondary school enrollmentrate; real investment’s share of GDP refers to investment in physicalcapital.

After combining the above variables from various data sources, weobtain an unbalanced panel data set of 46 countries over the 1970]89period. The Data Appendix gives the list of countries included.

3.2. Regression ResultsŽ . Ž .We estimate regression 3.1 using i three country groupings}the full

Ž . Ž .sample world , developing and developed country samples; ii five- andŽ .ten-year average data, and iii with and without the control variables. Our

Ž .baseline regression includes the first five regressors in 3.1 : a constant,average tax rate, fiscal decentralization, country fixed effects, and timefixed-effects. We then look at the sign and significance of the coefficienton the fiscal decentralization variable as we sequentially add the controlvariables across the three country groupings. These regressions, therefore,provide a rich set of sensitivity analyses regarding the possible relationshipbetween fiscal decentralization and growth. We have summarized all theseregressions in two tables.

Tables 1 and 2 show that there is a negative relationship between fiscaldecentralization and economic growth for five- and ten-year intervals inthe world and developing country samples; the point estimate is statisti-cally significant in a one-tail test at the 5% and 10% level for the worldand the developing country samples respectively. For the world sample, the

Ž . Ž .t-ratios are y2.00 Table 1; column 5 and y1.85 Table 2; column 5Ž .while the t-ratios for developing countries are y1.48 Table 1; column 5

Ž .and y1.72 Table 2; column 5 . The one-tail test is a suitable one becausethe hypothesis that fiscal decentralization leads to higher growth predicts apositive relationship. Clearly a negative relation is a rejection of thishypothesis at the stated significance levels. However, if we were agnosticand maintained a hypothesis of no relation, then we should employ atwo-tail test. In such a case we would not be able to reject the hypothesisof no relation at the same significance levels.

With either the one- or two-tail test, the point estimate of fiscaldecentralization is similar for both samples in the two tables. For example,

Ža 10 percentage point increase in fiscal decentralization well within its.standard deviation of 18% in the world and developing country samples is

associated with a reduction in the growth rate of 0.7]0.8 percentagepoints. As a benchmark comparison, consider another growth-reducingpolicy experiment: an equivalent reduction in the investment-GDP ratio of10 percentage points will lead to a much larger decline in the growth rateŽ2.3]3.2 percentage points in the world and developing country samples,

.respectively .

DAVOODI AND ZOU250

TABLE 1Five-Year Averages

Dep. Var: Per Capita GDP GrowthWorld sample

Ž . Ž . Ž . Ž . Ž .Indep. var. 1 2 3 4 5

Constant 1.53 4.41 5.53 48.78 51.95Ž . Ž . Ž . Ž . Ž .0.69 1.55 1.80 3.60 3.98

Average tax rate y0.04 y0.06 y0.04 y0.06 y0.01Ž . Ž . Ž . Ž . Ž .y0.58 y0.85 y0.49 y0.81 y0.14

Fiscal y0.07 y0.07 y0.06 y0.06 y0.08Ž . Ž . Ž . Ž . Ž .decentralization y1.86 y1.85 y1.47 y1.56 y2.00

Dummy for 1975]79 y1.09 y1.21 y1.01 y0.13 y0.17Ž . Ž . Ž . Ž . Ž .y2.02 y2.24 y1.58 y0.20 y0.26

Dummy for 1980]84 y2.34 y2.54 y2.45 y0.77 y0.60Ž . Ž . Ž . Ž . Ž .y4.00 y4.28 y3.23 y0.87 y0.71

Dummy for 1985]89 y1.19 y1.45 y0.98 0.99 1.39Ž . Ž . Ž . Ž . Ž .y1.91 y2.26 y1.09 0.95 1.36

Population growth y1.14 y1.43 y1.93 y2.91Ž . Ž . Ž . Ž .y1.61 y1.90 y2.63 y3.73

Initial human y0.04 y0.04 y0.06Ž . Ž . Ž .capital y1.05 y1.25 y1.68

Initial per capita y5.86 y6.17Ž . Ž .GDP y3.27 y3.58

Investment share 0.23Ž .of GDP 2.95

R-square 0.53 0.54 0.53 0.58 0.62Obs 158 157 145 145 145No. of countries 46 45 43 43 43F value 2.40 2.48 2.22 2.61 2.94Prob ) F 0.0001 0.0001 0.0005 0.0001 0.0001

Developed countries

Ž . Ž . Ž . Ž . Ž .Indep. var. 1 2 3 4 5

Constant 3.42 3.55 2.91 104.59 97.73Ž . Ž . Ž . Ž . Ž .1.07 1.01 0.69 4.89 4.57


Fiscal 0.01 0.01 0.02 0.04 0.04Ž . Ž . Ž . Ž . Ž .decentralization 0.18 0.16 0.42 0.77 0.77






TABLE 1Ž .Continued

Ž . Ž . Ž . Ž . Ž .Indep. var. 1 2 3 4 5

Initial human y0.01 y0.01 y0.01Ž . Ž . Ž .Capital y0.25 y0.57 y0.72




Developing countries

Ž . Ž . Ž . Ž . Ž .Indep. var. 1 2 3 4 5

Constant 3.46 8.40 9.75 63.76 69.64Ž . Ž . Ž . Ž . Ž .1.14 2.06 2.14 3.25 3.76




Dummy for 1980]84 y2.46 y2.69 y1.94 0.10 y0.34Ž . Ž . Ž . Ž . Ž .y2.55 y2.81 y1.42 0.07 y0.25







Note: t-statistics are in parentheses. All regressions include country-specific dummies.

DAVOODI AND ZOU252

TABLE 2Ten-Year Averages

Dep. Var: Per Capita GDP GrowthWorld sample

Ž . Ž . Ž . Ž . Ž .Indep. var. 1 2 3 4 5

Constant 1.51 2.52 5.14 60.76 60.50Ž . Ž . Ž . Ž . Ž .0.80 0.89 1.48 4.56 4.51








R-square 0.75 0.75 0.75 0.84 0.85Obs 87 86 82 82 82No. of countries 46 45 44 44 44F values 2.38 2.33 2.12 3.53 2.42Prob ) F 0.0033 0.0043 0.0128 0.0002 0.0002

Developed countries

Ž . Ž . Ž . Ž . Ž .Indep. var. 1 2 3 4 5

Constant y0.27 y3.15 y2.13 80.21 78.15Ž . Ž . Ž . Ž . Ž .y0.05 y0.56 y0.32 5.73 5.14

Average tax rate 0.09 0.13 0.14 0.11 0.10Ž . Ž . Ž . Ž . Ž .0.85 1.17 1.13 1.73 1.60

Fiscal y0.04 y0.02 y0.01 0.05 0.05Ž . Ž . Ž . Ž . Ž .decentralization y0.52 y0.28 y0.16 1.37 1.33


Population growth 1.14 y0.11 0.06 0.05Ž . Ž . Ž . Ž .1.11 y0.07 0.08 0.06



Ž .Investment share 0.03Ž .of GDP 0.46


TABLE 2Ž .Continued

Ž . Ž . Ž . Ž . Ž .Indep. var. 1 2 3 4 5

R square 0.63 0.66 0.67 0.92 0.93Obs 38 38 36 36 36No. of countries 19 19 19 19 19F value 1.31 1.323 1.083 5.647 5.04Prob ) F 0.2937 0.2922 0.4596 0.0024 0.0055

Developing countries

Ž . Ž . Ž . Ž . Ž .Indep. var. 1 2 3 4 5

Constant 3.79 5.27 10.50 58.38 58.60Ž . Ž . Ž . Ž . Ž .1.58 1.47 2.35 3.53 3.46



Dummy for 1980]89 y1.56 y1.64 0.36 y1.36 1.34Ž . Ž . Ž . Ž . Ž .y2.58 y2.59 y0.37 1.39 1.34






Note: t-statistics are in parentheses. All regressions include country-specific dummies.

As for other explanatory variables in the regressions, the average taxrate is negatively related to growth and quantitatively more significant fordeveloping countries. This finding perhaps reflects the differential effectsof distortionary taxation between the two sets of countries. Of the four

Ž .other control variables, only human capital has the wrong negative sign;growth is higher in countries with a higher investment rate, lower popula-tion growth and lower initial per capita GDP. The negative coefficient on

DAVOODI AND ZOU254

initial per capita GDP indicates the conditional convergence found inmany previous studies, i.e., other things being equal, countries that startpoorer tend to grow faster.

Ž .To sum up, regression results in Tables 1 and 2 show that i there is norelationship between fiscal decentralization and growth in developed coun-

Ž .tries, and ii the negative relationship between fiscal decentralization andgrowth is limited to the developing country sample, which seems to bedriving the same results in the pooled world sample. To explain thedifference, we note that in our panel data there is much more cross-coun-try variation in growth and fiscal decentralization for developing countriesthan developed countries. The standard deviation of per capita outputgrowth in developing countries is three times as high as that of developedcountries; and the spread between the most and the least fiscally-de-centralized developing country is 1.4 times as high as the spread indeveloped countries. Developed countries are simply too homogeneousvis-a-vis developing countries, leaving hardly any cross-country variation infiscal decentralization to be linked systematically to cross-country differ-ences in growth. Therefore, when the data for developing countries arepooled with developed countries, the variation in the former dominatesthat in the latter, producing a result for the world sample that closelyresembles the developing country sample.

3.3. Some Explanations of the Negati e Effect of Fiscal Decentralization onGrowth in De¨eloping Countries

How can one explain the negative impact of fiscal decentralization oneconomic growth for developing countries? We offer several explanations.First, the composition of government spending may explain the negativefinding. The decentralization measure in this study does not tell us what asubnational government buys; it does not distinguish between current

Ž .spending e.g., wages and salaries and capital spending; nor does itdistinguish spending on welfare and social security from infrastructurespending. The conventional wisdom points towards positive growth effectsof capital and infrastructure spending and negative growth effects ofwelfare and current spending. Excessive spending by subnational govern-ments on the wrong expenditure items can lead to lower growth even if theexpenditure assignment is optimal. Second, lower growth can result fromthe wrong revenue assignment among various levels of government. Forexample, subnational governments may be raising revenues using a taxinstrument which should have been used by the central government. Third,the efficiency gains from fiscal decentralization, perhaps the strongest


argument in its favor, may not materialize for developing countries sincerevenue collection and expenditure decisions by local governments maystill be constrained by the central government. Fourth, in practice localgovernments may not be responsive to local citizens’ preferences andneeds. This can occur when local officials are not elected by local citizensand when local citizens may be too poor to ‘‘vote with their feet.’’

4. CONCLUSIONS

In this paper we have provided a simple endogenous growth modelshowing how the degree of fiscal decentralization affects the growth rate ofthe economy. We used a cross-country panel data set of 46 developed anddeveloping countries over the 1970]89 period to investigate whether fiscaldecentralization has any growth impact. From our sample, developedcountries are on average more decentralized than developing countriesŽ . Ž33% vs. 20% and tend to have a higher per capita GDP growth rate 2%

.vs. 1.6% . But can one conclude that there is a positive relationshipbetween decentralization and growth? Given other determinants of growth,we find a negative relationship for developing countries and the world, andnone for developed countries. The point estimate for the developingcountry sample is significant at the 10% level in a one-tail test, which is anappropriate test under the null hypothesis that fiscal decentralization leadsto higher growth. Therefore, on the basis of this test and the significancelevel we clearly reject the hypothesis of a positive relationship betweenfiscal decentralization and growth.

Finally, we want to draw attention to one major limitation of thispreliminary study. Our measure of fiscal decentralization, which is thesubnational government share of total government expenditure, may notreflect the subnational governments’ autonomy in expenditure decision-

w xmaking. As Musgrave 10 has pointed out, subnational governments thatact as administrative agents of national governments do not necessarilyreflect true expenditure decentralization. Further work will look into otherrelated issues of fiscal decentralization such as local autonomy, localrevenue collection, the composition of local spending, and intergovern-mental transfers.4

4 w xSome analyses in a dynamic framework can be found in Zou 18, 19 .

DAVOODI AND ZOU256

APPENDIX

List of the 46 Countries in the Sample

Argentina ZimbabweChile BelgiumCzechoslovakia, Former DenmarkFinland FranceHungary IrelandIceland LuxembourgIndia NetherlandsIndonesia PortugalIran, Islamic Republic of United KingdomIsrael PolandItaly ChinaKenya AustraliaMalawi AustriaMalaysia BoliviaMexico BrazilNetherlands Antilles CanadaNorway ColombiaParaguay South AfricaPhilippines SwitzerlandRomania United StatesSweden GermanyThailand SpainUruguay Yugoslavia, Former Fed. Rep.

REFERENCES

1. R. Bahl and J. F. Linn, ‘‘Urban Public Finance in Developing Countries,’’ New York,Ž .Oxford Univer. Press 1992 .

Ž .2. R. Barro and X. Sala-i-Martin, ‘‘Economic Growth,’’ McGraw-Hill, New York 1995 .3. R. Barro, Government spending in a simple model of endogenous growth, Journal of

Ž .Political Economy, 98, S103]S125 1990 .4. R. Bird and C. Wallich, ‘‘Fiscal decentralization and intergovernmental relations in

transition economies: Towards a systematic framework of analysis,’’ Country Eco-Ž .nomics Department Working Paper, World Bank, Washington D.C. 1993 .

5. H. Davoodi, D. Xie, and H. Zou, ‘‘Fiscal Decentralization and Economic Growth in theŽ .United States,’’ Mimeo, Policy Research Department, World Bank 1995 .

6. S. Devarajan, V. Swaroop, and H. Zou, The composition of public expenditure andŽ .economic growth, Journal of Monetary Economics, 37, 313]344 1996 .

7. W. Dillinger, Decentralization and its implications for urban service delivery,’’ UrbanManagement and Municipal Finance Discussion Paper No. 16, The World Bank,

Ž .Washington D.C. 1994 .


8. International Monetary Fund, ‘‘Government Finance Statistics,’’ Washington, DC.9. R. Levine and D. Renelt, A sensitivity analysis of cross-country growth regressions,

Ž .American Economic Re¨iew, 82, 942]963 1992 .10. R. Musgrave, ‘‘The Theory of Public Finance, A Study in Public Economy,’’ McGraw-Hill,

Ž .New York 1959 .Ž .11. W. Oates, ‘‘Fiscal Federalism,’’ Harcourt Brace Jovanovich, New York 1972 .

12. W. Oates, Searching for Leviathan: An empirical analysis, American Economic Re¨iew,Ž .75, 748]757 1985 .

13. W. Oates, Fiscal decentralization and economic development, National Tax Journal,Ž .XLVI, 237]243 1993 .

Ž .14. R. Summers and A. Heston, Summers]Heston Database, version 5.6a 1995 .15. C. Tiebout, A pure theory of local expenditures, Journal of Political Economy, 64,

Ž .416]424 1956 .16. The World Bank, ‘‘World Bank Economic and Social Database,’’ Washington, DC.17. T. Zhang and H. Zou, ‘‘Fiscal Decentralization, Public Spending and Economic Growth

in China,’’ Journal of Public Economics, in press.18. H. Zou, Dynamic effects of federal grants on local spending, Journal of Urban Economics,

Ž .36, 98]115 1994 .19. H. Zou, Taxes, federal grants, local public spending and growth, Journal of Urban

Ž .Economics, 39, 303]317 1996 .

fiscal decentralization and economic growth: a cross-country study

Documents